U.S. patent application number 12/358721 was filed with the patent office on 2010-07-29 for spin device.
Invention is credited to Gerhard Poeppel, Werner Robl, Hans-Joerg Timme.
Application Number | 20100188905 12/358721 |
Document ID | / |
Family ID | 42317623 |
Filed Date | 2010-07-29 |
United States Patent
Application |
20100188905 |
Kind Code |
A1 |
Poeppel; Gerhard ; et
al. |
July 29, 2010 |
Spin Device
Abstract
According to an embodiment of the present invention, a spin
device includes an intermediate semiconductor region arranged
between a first terminal and a second terminal, wherein the first
terminal is adapted to provide a current having a first degree of
spin polarization to the intermediate semiconductor region, and
wherein the second terminal is adapted to output the current having
a second degree of spin polarization. The spin device further
includes a spin selective scattering structure abutting the
intermediate semiconductor region, the spin selective scattering
structure being adapted such that the first degree of spin
polarization is altered to be the second degree, wherein the spin
selective scattering structure comprises a control electrode being
electrically insulated from the intermediate semiconductor region,
and wherein the control electrode is adapted to apply an electrical
field perpendicular to a direction of the current through the
intermediate semiconductor region to control a magnitude of the
current.
Inventors: |
Poeppel; Gerhard;
(Regensburg, DE) ; Timme; Hans-Joerg; (Ottobrunn,
DE) ; Robl; Werner; (Regensburg, DE) |
Correspondence
Address: |
SLATER & MATSIL LLP
17950 PRESTON ROAD, SUITE 1000
DALLAS
TX
75252
US
|
Family ID: |
42317623 |
Appl. No.: |
12/358721 |
Filed: |
January 23, 2009 |
Current U.S.
Class: |
365/185.28 ;
257/24; 257/E29.168 |
Current CPC
Class: |
B82Y 25/00 20130101;
H01F 10/3254 20130101; H01L 29/66984 20130101 |
Class at
Publication: |
365/185.28 ;
257/24; 257/E29.168 |
International
Class: |
G11C 15/02 20060101
G11C015/02; H01L 29/66 20060101 H01L029/66 |
Claims
1. A spin device comprising: an intermediate semiconductor region
arranged between a first terminal and a second terminal, wherein
the first terminal is adapted to provide a current having a first
degree of spin polarization to the intermediate semiconductor
region, wherein the second terminal is adapted to output the
current, having passed the intermediate semiconductor region,
having a second degree of spin polarization; a spin selective
scattering structure abutting the intermediate semiconductor
region, the spin selective scattering structure being adapted such
that the first degree of spin polarization is altered to be the
second degree of spin polarization; and a control electrode
electrically insulated from the intermediate semiconductor region
and adapted to apply an electrical field to the intermediate
semiconductor region to control a magnitude of the current, the
electrical field having a main field component perpendicular to a
direction of the current through the intermediate semiconductor
region.
2. The spin device according to claim 1, wherein the spin selective
scattering structure comprises a ferromagnetic layer, wherein an
interface between the spin selective scattering structure and the
intermediate semiconductor region is formed such that a scattering
property of a charge carrier of the intermediate semiconductor
region depends on a spin orientation of the charge carrier.
3. The spin device according to claim 2, wherein the ferromagnetic
layer is a conducting ferromagnetic layer on an insulating layer,
and wherein the insulating layer is formed abutting the
intermediate semiconductor region to form the interface.
4. The spin device according to claim 3, wherein the insulating
layer comprises a thickness such that the scattering property of
the charge carrier of the intermediate semiconductor region depends
on the spin orientation of the charge carrier.
5. The spin device according to claim 3, wherein the spin selective
scattering structure comprises an insulating layer, wherein the
conducting ferromagnetic layer and the intermediate semiconductor
region are adapted such that charge carriers in the intermediate
semiconductor region are attractable towards the interface by
applying a voltage to the ferromagnetic layer during operation of
the spin device.
6. The spin device according to claim 1, wherein the spin selective
scattering structure further comprises a biasing electrode that is
electrically insulated from the intermediate semiconductor region,
wherein the biasing electrode and the intermediate semiconductor
region are adapted such that charge carriers in the intermediate
semiconductor region are attractable towards an interface by
applying a voltage to the biasing electrode during operation of the
spin device.
7. The spin device according to claim 1, wherein the control
electrode is electrically insulated from the intermediate
semiconductor region by a further insulating layer, wherein the
further insulating layer is arranged such that a further interface
is formed between the further insulating layer and a part of the
intermediate semiconductor region.
8. The spin device according to claim 1, wherein the spin device
comprises a floating gate electrode electrically insulated from the
control electrode by a tunneling insulating layer such that a
charge of the floating gate electrode is changeable by tunneling
from the control electrode and the charged state of the floating
gate electrode contributes to the electrical field perpendicular to
the direction of the current.
9. The spin device according to claim 1, wherein the intermediate
semiconductor region comprises silicon.
10. A spin device comprising: an intermediate semiconductor region
arranged between a first terminal and a second terminal, wherein
the first terminal is adapted to provide a current having a first
degree of spin polarization to the intermediate semiconductor
region, wherein the second terminal is adapted to output the
current, having passed the intermediate semiconductor region,
having a second degree of spin polarization; and a spin selective
scattering structure abutting the intermediate semiconductor
region, the s spin selective scattering structure being adapted
such that the first degree of spin polarization is altered to be
the second degree of spin polarization, wherein the spin selective
scattering structure comprises a control electrode that is
electrically insulated from the intermediate semiconductor region,
the control electrode adapted to apply an electrical field to the
intermediate semiconductor region to control a magnitude of the
current, the electrical field having a main field component
perpendicular to a direction of the current through the
intermediate semiconductor region.
11. The spin device according to claim 10, wherein the spin
selective scattering structure comprises an insulating layer
abutting the intermediate semiconductor region such that an
interface between the spin selective scattering structure and the
intermediate semiconductor region is formed, wherein the control
electrode comprises a ferromagnetic material or a ferromagnetic
compound, and wherein the control electrode is formed on the
insulating layer.
12. The spin device according to claim 11, wherein the insulating
layer comprises a thickness such that a scattering property of
charge carriers in the intermediate semiconductor region depend on
a spin orientation of the charge carriers.
13. The spin device according to claim 10, wherein the spin
selective scattering structure further comprises a ferromagnetic
layer, wherein an interface between the spin selective scattering
structure and the intermediate semiconductor region is formed such
that a scattering property of charge carriers in the intermediate
semiconductor region depends on a spin orientation of the charge
carriers.
14. The spin device according to claim 13, wherein the spin
selective scattering structure comprises an insulating layer,
wherein the ferromagnetic layer comprises a conducting
ferromagnetic material or a conducting ferromagnetic compound,
wherein the ferromagnetic layer is formed on an insulating layer,
wherein the insulating layer is formed abutting the intermediate
semiconductor region to form the interface, and wherein the
insulating layer comprises a thickness such that the scattering
property of the charge carriers depends on the spin orientation of
the charge carriers.
15. The spin device according to claim 14, wherein the control
electrode is electrically insulated from the ferromagnetic layer by
a tunneling insulating layer such that the ferromagnetic layer is a
floating electrode, a charge state of the ferromagnetic layer is
changeable by tunneling of charge carriers from the control
electrode and the charge state of the ferromagnetic layer
contributes to the electrical field perpendicular to the direction
of the current.
16. The spin device according to claim 14, further comprising a
floating gate electrode electrically insulated from the control
electrode such that a charge state of the floating gate electrode
is changeable by tunneling from the control electrode and the
charge state of the floating gate electrode contributes to the
electrical field perpendicular to the direction of the current.
17. The spin device according to claim 10, wherein the control
electrode and intermediate semiconductor region are adapted such
that charge carriers in the intermediate semiconductor region are
attractable towards an interface between the spin selective
scattering structure and the intermediate semiconductor region by
applying a voltage to the control electrode during operation of the
spin device.
18. The spin device according to claim 10, wherein the spin device
comprises a field effect transistor structure with the intermediate
semiconductor region comprising a semiconductor region in which a
channel area is formed during operation, with the first terminal
being a source terminal, with the second terminal being a drain
terminal, and with the control electrode being a gate
electrode.
19. The spin device according to claim 10, wherein the intermediate
semiconductor region comprises silicon.
20. A method for providing a current having a second degree of spin
polarization based on a current having a first degree of spin
polarization, the method comprising: spin selective scattering of
charge carriers of the current such that the first degree of spin
polarization is altered to be the second degree of spin
polarization; and controlling a magnitude of the current by
applying an electrical field having a main component perpendicular
to a direction of the current.
21. The method according to claim 20, wherein the spin selective
scattering of the charge carriers comprises providing an interface
such that a scattering property of the charge carriers depends on a
spin orientation of the charge carriers.
22. The method according to claim 21, wherein the spin selective
scattering of the charge carriers comprises increasing an
interaction of the charge carriers with the interface by applying a
biasing electrical field such that charge carriers are attracted
towards the interface.
23. The method according to claim 20, wherein controlling the
magnitude of the current comprises supplying a voltage to a control
electrode.
24. The method according to claim 20, wherein controlling the
magnitude of the current comprises changing a charged state of a
floating gate electrode, the charged state of the floating gate
electrode contributing to the electrical field perpendicular to the
direction of the current.
25. A spin device comprising: an intermediate semiconductor region
arranged between a first terminal and a second terminal, wherein
the first terminal is adapted to provide a current having a first
degree of spin polarization to the intermediate semiconductor
region, wherein the second terminal is adapted to output the
current, having passed the intermediate semiconductor region,
having a second degree of spin polarization; and a spin selective
scattering structure abutting the intermediate semiconductor
region, the spin selective scattering structure being adapted such
that the first degree of spin polarization is altered to be the
second degree of spin polarization, wherein the spin selective
scattering structure further comprises a floating control
electrode, the floating control electrode being electrically
insulated from the intermediate semiconductor region, a control
electrode being electrically insulated from the intermediate
semiconductor region and being electrically insulated from the
floating control electrode by a tunneling insulating layer, wherein
the tunneling insulating layer is adapted to allow a tunneling of
electrical charges from the control electrode to the floating
control electrode, and wherein the control electrode and the
floating control electrode are adapted to apply an electrical field
to the intermediate semiconductor region to control the current,
the electrical field having a main component perpendicular to a
direction of the current through the intermediate semiconductor
region.
26. The spin device according to claim 25, wherein the control
electrode and the intermediate semiconductor region are adapted
such that charge carriers in the intermediate semiconductor region
are attractable towards an interface formed between the spin
selective scattering structure and the intermediate semiconductor
region by applying a voltage to the control electrode during
operation of the spin device.
27. A spin device comprising: a first intermediate semiconductor
region arranged between a first terminal and a second terminal,
wherein the first terminal is adapted to provide a first current
having a first degree of spin polarization to the first
intermediate semiconductor region, a first spin selective
scattering structure abutting the first intermediate semiconductor
region, the first selective spin scattering structure being adapted
such that the first degree of spin polarization is altered to be a
second degree of spin polarization, a second intermediate
semiconductor region arranged between the first terminal and a
third terminal, wherein the first terminal is further adapted to
provide a second current having the first degree of spin
polarization to the second intermediate semiconductor region, a
second spin selective scattering structure abutting the second
intermediate semiconductor region, the second spin selective
scattering structure being adapted such that the first degree of
spin polarization is altered to be a third degree of spin
polarization, wherein the second terminal is adapted to output the
first current; wherein the third terminal is adapted to output the
second current; and wherein the first spin selective scattering
structure comprises a first control electrode being electrically
insulated from the first intermediate semiconductor region, the
first control electrode being adapted to apply a first electrical
field to the first intermediate semiconductor region to control a
magnitude of the first current, the first electrical field having a
main component perpendicular to a direction of the first current
through the first intermediate semiconductor region; wherein the
second spin selective scattering structure comprises a second
control electrode being electrically insulated from the second
intermediate semiconductor region, the second control electrode
being adapted to apply a second electrical field to the second
intermediate semiconductor region to control a magnitude of the
second current, the second electrical field having a main component
perpendicular to a direction of the second current through the
second intermediate semiconductor region.
Description
BACKGROUND
[0001] In conventional electrical engineering as well as in
conventional electrical and electronic devices, the principle
physical quantity of charge carriers employed is their electrical
charge. The behavior of the charge carriers, such as electrons,
holes, or other quasiparticle excitations, in these devices is
mainly influenced by electrical fields, for instance by applying a
voltage, or by a magnetic field, which interacts with moving charge
carriers by virtue of the Lorentz force.
[0002] However, charge carriers often comprise further physical
quantities which may be exploited in the framework of appropriate
devices. Since such a further physical property of the charge
carriers, apart from their electrical charge, may additionally be
employed, a further degree of interaction with the charge carriers
may be used for the design and construction of novel devices.
[0003] Such an additional physical quantity is the spin of charge
carriers which is a quantum physical property of electrons, holes
and other quasiparticle excitations.
[0004] However, to create devices exploiting spin-based physical
effects, a source for at least partially spin-polarized currents or
charge carriers, as well as a detector for detecting the spin state
of charge carriers or a current of charge carriers, is needed to
provide and/or to detect interactions with this physical quantity.
Needless to say, such a source or a detector should also be
compatible with the underlying technology.
[0005] While a wide range of metallic elements, compounds and
materials exists, which may be utilized to provide or to detect an
at least partially spin-polarized current, especially in the field
of semiconductor technology, such a source for an at least
spin-polarized current or an appropriate detector is still not
satisfactorily implemented. In metallic systems, ferromagnetic
metals, compounds, or elements such as iron (Fe), cobalt (Co), or
nickel (Ni) may be used. However in semiconductor systems and
devices this need is still not satisfied.
BRIEF DESCRIPTION OF THE DRAWINGS
[0006] Embodiments according to the present invention will be
described hereinafter making reference to the appended
drawings.
[0007] FIG. 1a shows a cross-sectional view of a spin device
according to an embodiment of the present invention;
[0008] FIG. 1b shows a cross-sectional view of a spin device
according to a further embodiment of the present invention;
[0009] FIG. 2a shows a top view of a spin device according to an
embodiment of the present invention;
[0010] FIG. 2b shows a cross-sectional view of the spin device
shown in FIG. 2a;
[0011] FIG. 3 shows a cross-sectional view of a spin device
according to a further embodiment of the present invention;
[0012] FIGS. 4a and 4b illustrate the operational principles of a
spin selective scattering structure as employed in embodiments
according to the present invention;
[0013] FIG. 5a shows a cross-sectional view of a spin device
according to an embodiment of the present invention;
[0014] FIG. 5b shows a cross-sectional view of a spin device
according to an embodiment of the present invention;
[0015] FIG. 6a shows a cross-sectional view of a spin device
according to an embodiment of the present invention comprising a
floating gate electrode;
[0016] FIG. 6b shows a cross-sectional view of a further spin
device comprising a floating gate electrode according to an
embodiment of the present invention;
[0017] FIGS. 7a and 7b show cross-sectional views of spin devices
according to an embodiment of the present invention with separated
spin scattering structures and control electrodes;
[0018] FIGS. 8a and 8b show cross-sectional views of spin devices
according to an embodiment of the present invention with a biasing
electrode;
[0019] FIGS. 9a and 9b show cross-sectional views of a spin device
according to an embodiment of the present invention comprising a
floating gate electrode;
[0020] FIGS. 10a and 10b shows cross-sectional views of spin
devices according to embodiments of the present invention
comprising floating gate electrodes and biasing electrodes;
[0021] FIGS. 11a and 11b show top views of a spin device according
to an embodiment of the present invention in the form of a lateral
device;
[0022] FIG. 11c shows a cross-sectional view along a line shown in
FIG. 11b through the spin device according to an embodiment of the
present invention;
[0023] FIG. 12a shows a top view of two spin devices according to
embodiments of the present invention in the form of lateral
devices;
[0024] FIG. 12b shows a cross-sectional view of the devices shown
in FIG. 12a;
[0025] FIG. 13a shows schematically in perspective a spin device
according to an embodiment of the present invention with one spin
injector;
[0026] FIG. 13b shows schematically in perspective a spin device
according to an embodiment of the present invention with two spin
injectors;
[0027] FIG. 14 shows schematically in perspective a spin device
comprising floating gate electrodes according to an embodiment of
the present invention; and
[0028] FIGS. 15a and 15b show simplified block diagrams of spin
devices according to embodiments of the present invention used as a
detector for detecting a spin polarization.
DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS
[0029] While in conventional electrical and electronic devices the
principle physical quantity exploited is the electrical charge,
novel devices arise in which further physical quantities such as
the quantum mechanical spin are additionally or alternatively
exploited. While in conventional devices the charge of charge
carriers is exploited by applying electrical fields (e.g., by
applying a voltage) or by applying a magnetic field which interacts
with a moving charge carrier by virtue of the Lorentz force,
additionally exploiting the quantum mechanical spin as a further
physical quantity significantly increases the degree of freedom to
interact with the charged carriers. The spin of a charge carrier,
but also of other uncharged particles, is often accompanied by a
magnetic moment. Moreover, the spin of the charge carrier, or a
non-charged particle, may also influence the statistical behavior
of the respective particles.
[0030] In the following, emphasis will be laid on charged particles
or quasiparticles, such as electrons and holes which allow not only
a manipulation of the spin, but also an interaction based on their
charged state, for instance, by applying electrical or magnetic
fields.
[0031] Exploiting not only the electrical properties of the charge
carriers, but also the spin, opens the possibility of implementing
so-called spintronic devices or spin electronic devices. The
technical field of such devices is sometimes referred to as
spintronics.
[0032] Spintronics, however, relies on providing charge carriers
with an at least partially spin-polarized current and detecting the
same. Therefore, a need exists for spin devices capable of
providing an at least partially spin-polarized current, spin
devices capable of changing a physical, or an electrical property
based on a spin polarization of a current provided to such a spin
device to act as a spin detector and other spin-related devices.
Spin devices according to the present invention may, therefore, be
used as switchable spin injection devices, which are sometimes also
referred to as switchable spinjectors, spin storage elements, spin
control elements, spin detectors, and other spin-related devices
for future logic or memory products based on spintronics.
[0033] Selective spin injection is, in many applications, an
important pre-condition for spin devices in general and for the
construction of spin circuitries, which come up as a new kind of
electronic technology.
[0034] Spin injection is a field of research today. Some of the
available ideas comprise a spin injection with a fixed orientation
from a metallic magnetic layer (e.g., ferromagnetic layer of iron
(Fe), nickel (Ni) or cobalt (Co)) with a fixed orientation to force
at least a small fraction of electrons with a certain kind of spin
into a semiconductor structure. However, at the interface between
the metallic layer and the semiconductor region, the obtainable
spin polarization is severely degraded. Moreover, problems may
arise when different sources for different spin orientations are
demanded in close vicinity to one another.
[0035] Among other applications, which will further be outlined
below, the following embodiments according to an embodiment of the
present invention offer, for instance, the possibility to easily
achieve a selective spin injection in semiconducting devices.
[0036] Before describing the physical processes and directions
leading to spin injection in more detail, first with respect to
FIGS. 1a to 3, embodiments according to the present invention will
be described. Spin devices according to the present invention
comprise a spin selective scattering structure, a possible
implementation of which will be outlined in the context of FIGS. 4a
and 4b.
[0037] It should be noted that in the following a spin-polarized
current and an at least partially spin-polarized current will be
synonymously used. A spin-polarized current is to be understood as
a current in which a relative number of charge carriers, having a
first spin orientation, is different from a number of charge
carriers having a second spin orientation different from the first
orientation. For instance, when two-thirds of the charge carriers
of a current comprise the first spin orientation, whereas only
one-third of the charge carriers of the current comprise the
second, different spin orientation, the spin-polarization or degree
of spin-polarization is 2:1 or 2/3.
[0038] Sometimes, the spin polarization is also expressed in terms
of a deviation from an unpolarized current. An unpolarized current
comprises, as a statistical average, an equal number of charge
carriers of either spin orientation. As a consequence, the ratio of
the number of charge carriers is 1:1 or 1/2. Therefore, a spin
polarization or a degree of spin polarization may be equally
well-expressed in terms of the deviation with respect to the spin
polarization of an unpolarized current. In the case of the example
above, the spin polarization or degree of spin polarization may be
equally well-expressed as 16.67% (=1/6=2/3-1/2). Based on the
second definition of the spin polarization, a fully spin-polarized
current, in which all charge carriers have the same spin
orientation, has spin polarization of 100% (=1=1-0), whereas an
unpolarized current has a spin polarization of 0% (=0=1/2-1/2).
[0039] To distinguish between the two definitions of spin
polarization, in the following a spin polarization expressed in
terms of a percentage value refer to deviation of a spin
polarization with respect to the spin contributions of an
unpolarized current, whereas a spin polarization expressed in terms
of a ratio is directly related to the number of respective charge
carriers.
[0040] Moreover, unless stated otherwise in the following, devices
according to embodiments of the present invention will be described
based on semiconductor implementations. Such spin devices may be
implemented based on electrons or based on holes as charge
carriers. Therefore, unless stated otherwise, whenever charge
carriers are referred to, an implementation based on electrons or
holes may in principle equally well be used. However, some
differences may arise, since holes and electrons may comprise
different parameters and physical quantities.
[0041] As a last comment before describing a first embodiment
according to the present invention, it should be noted that in many
implementations, the charge carriers are spin 1/2-particles which
means that these particles may only acquire two distinguished spin
states, which are denoted as +1/2 and -1/2, respectively. In some
devices, for instance, when a specific interaction (e.g., spin
orbit interaction) becomes the dominating interaction, the spin may
be replaced by an effective quantum number indicating an overall
momentum, which may lead to the charge carriers acquiring effective
momentum quantum numbers being different from the previously
mentioned two states. However, in most cases, considering the
charge carriers as being capable of only acquiring the two
previously mentioned spin states +1/2, -1/2, is not only a
sufficient description to understand the electrical behavior of the
spin devices, but also an accurate description in terms of the
underlying physics.
[0042] FIG. 1a shows a cross-sectional view of a spin device 100
according to an embodiment of the present invention based on a
semiconducting substrate 110. The substrate 110 comprises an
intermediate semiconductor region 120 which is arranged between a
first terminal 130 and a second terminal 140. The terminals 130,
140 may, for instance, be implemented as doping areas. The first
terminal 130 may be coupled to a first contact 150, which is
schematically shown in FIG. 1a. Accordingly, the second terminal
140 may be coupled, in a similar manner, to a second contact
160.
[0043] The spin device 100 further comprises, on top of and
abutting the intermediate semiconductor region 120, a spin
selective scattering structure 170, such that an interface 180 is
formed between the spin selective scattering structure 170 and the
intermediate semiconductor region 120. In the embodiment shown in
FIG. 1a, a control electrode 190 is deposited on top of the spin
selective scattering structure 170, which extends along the
cross-sectional view of FIG. 1a along the whole length of the spin
selective scattering structure 170. A control contact 200 may be
coupled to the control electrode 190 to allow an electrical contact
of the electrode 190.
[0044] Concerning the mode of operation of the spin device 100 as
shown in FIG. 1a, the first terminal 130 is adapted to provide a
current with a first degree of spin polarization to the
intermediate semiconductor region 120. Depending on the
circumstances of the operation and other parameters, the first
degree of spin polarization may, for instance, be that of an
unpolarized current, i.e. spin polarization 1:1 or 0%. Therefore,
charge carriers of the current are injected into the intermediate
semiconductor region 120 via the first contact 130. Here, the
charge carriers interact with the spin selective scattering
structure 170 via the interface 180. As a consequence, the charge
carriers are spin selectively scattered such that the degree of
spin-polarization of current changes. It changes in such a manner
that the degree of spin polarization of the current having passed
the intermediate semiconductor region 120 and provided by the
second terminal 140 comprises a second degree of spin-polarization,
which may be typically different from the first
spin-polarization.
[0045] However, it may happen that the first degree of spin
polarization is equal to the second degree of spin polarization.
This occurs, for instance, when the degree of the spin-polarization
of the current provided to the intermediate semiconductor region
120 is equal to an equilibrium distribution of the charge carriers
among the two spin orientations, after having traveled through the
intermediate semiconductor region 120 and interacting with the
interface 180, or the spin selective scattering structure 170. This
is a situation which may occur, for instance, in the case of the
spin device 100 operating as a spin detector.
[0046] It should be noted that the magnitude of the current is, to
a good approximation, not changed by the interaction of the charge
carriers with the spin selective scattering structure 170. The
distribution of the spin orientations between the two possible spin
orientations (in the case of a two-spin state system) or among the
number of available spin states is merely altered by the
interaction of the charge carriers with the selective scattering
structure 170. No charge carriers are typically trapped or removed
on a long term basis from the current flowing between the first and
second terminals 130, 140 by the spin selective scattering
structure 170. The amount of charge transported, and hence the
magnitude of the current, is not influenced by the spin selective
scattering structure 170.
[0047] Due to spin selective scattering processes, such as
processes comprising a spin flip, or an increased number of
scattering incidences for one spin orientation compared to the
number of interactions for the other spin orientation, this may
lead to the described alteration of the distribution of the charge
carriers among the available spin states.
[0048] In embodiments according to the present invention, the spin
selective scattering structure 170 may comprise an electrically
insulating layer to, for instance, prevent electrical short
circuits or other unwanted influences between the spin selective
scattering structure 170 and the intermediate semiconductor region
120.
[0049] The spin device 100 according to an embodiment of the
present invention is capable of controlling the magnitude of the
current by applying an electrical field having a main component
perpendicular to a direction 210 of the current through the
intermediate semiconductor region 120. In the following, for the
sake of simplicity only, electrical fields having a main component
perpendicular to an intermediate semiconductor region, a direction
of a current or a similar structure or direction will be referred
to as an electrical field being perpendicular to the respective
direction or structure. An electrical field being perpendicular to
such a structure or direction or having a main component
perpendicular to such a structure or direction is an electrical
field with a relevant component in terms of the underlying physics.
Hence, in some cases stray fields or other deviations from ideal
conditions may have to be considered not to have a main component
or being perpendicular. Electrical fields being applied in the
range of angles with respect to a normal up to 60.degree.,
45.degree., 30.degree. or 10.degree. may be considered as having a
main component perpendicular and, hence, being perpendicular,
depending on the implementation in mind.
[0050] Depending on the concrete implementation of a spin device
100, according to an embodiment of the present invention, the
physical effect employed may be different. For instance, the spin
device 100 may be implemented as a field effect transistor
(FET=Field Effect Transistor), such that the first terminal 130 and
the second terminal 140 may be implemented as a source terminal and
drain terminal, or vice versa. In this case, a channel or a channel
region is formed in at least a part of the intermediate
semiconductor region 120, the electrical properties of which are
controlled by the electrical field applied via the control
electrode 190, which may also be referred to as a gate electrode in
case of a FET. For instance, in the case of an implementation in
the form of an enhancement field effect transistor, by applying a
sufficiently strong electrical field, the channel region between
the first and second terminals 130, 140, is formed. In the case of
an n-channel field effect transistor, the intermediate
semiconductor region may be formed by a p-type semiconductor
region, while the first and second terminals 130, 140 may be formed
as n-type regions or wells.
[0051] Naturally, a spin device 100 based on such a FET layout may
also be implemented as a depletion FET, a p-channel FET or a
p-channel enhancement FET. In the case of a p-channel enhancement
FET, the intermediate semiconductor region 120 comprises an n-type
semiconductor region in which the appropriate p-channel region is
formed by applying the respective electrical field. In this case,
the first and second terminals 130, 140 are also p-type
semiconductor regions or wells. For the sake of simplicity only,
doping profiles and regions are not shown in the Figs.
[0052] However, in different embodiments according to the present
invention, the spin device 100 may also be formed on the basis of
different semiconductor devices. For instance, the intermediate
semiconductor region 120 may be based on a thin layer of
semiconductor material deposited on top of an insulating layer 220
which may optionally be implemented, as shown by the dashed lines
in FIG. 1a. Such a buried insulating layer 220 may, for instance,
be fabricated based on the available SOI-techniques (SOI=Silicon on
Insulator). Among the available techniques, the SIMOX-technique
(SIMOX=Separation by Implantation of OXygen) or the BESOI-technique
(BESOI=Bonded Etched-Back Silicon on Insulator) exist, to name but
two. Such an embodiment according to the present invention will be
described in more detail with respect to FIG. 2, which includes
FIG. 2a and 2b.
[0053] Therefore, embodiments according to the present invention in
the form of a spin device 100 are based on the finding that the
generation of a spin-polarized current or detection of a degree of
spin-polarization is achievable in a semiconductor structure
(intermediate semiconductor region 120), by providing spin
selective scattering structure 170 abutting the intermediate
semiconductor region 120 to eventually influence the degree of spin
polarization. The magnitude of the current is controllable via the
electrical fields applicable perpendicular to the direction 210 of
the current, which is directed from the first terminal 130 to the
second terminal 140. As a consequence, a spin device 100, according
to an embodiment of the present invention, may offer the
possibility of providing a spin-polarized current or to detect a
spin-polarized current in a semiconductor region in a controllable
manner.
[0054] A spin device 100 according to an embodiment of the present
invention may be employed as a detector or an injector for a
spin-polarized current. Hence, a spin device 100 according to an
embodiment of the present invention may eventually be comparable to
an optical polarization filter which may also be used as a
generator for polarized light in the framework of a light source,
or as a filter in the framework of the detector. Therefore, a spin
device 100 may also operate as a filter or a generator, as the
optical analogy shows. However, in contrast to the optical
polarization filter, it is further capable of controlling the
magnitude of the current.
[0055] With respect to FIG. 1a, it should be noted that it may be
possible to restrict the extension of the control electrode 190 to
a smaller area, compared to the extension of the control electrode
190, as shown in FIG. 1a. For instance, by restricting the control
electrode 190 to only the central region directly underneath the
control contact 200, the perpendicular electrical field induced
into the intermediate semiconductor region 120 may create a
modification of the potentials present in the intermediate
semiconductor region 120, so that the transport of the charge
carriers, and hence, the current, will be affected.
[0056] However, since the selective spin scattering effect is
caused by an interaction of the charge carriers with the interface
180 formed between the spin selective scattering structure 170 and
the intermediate semiconductor region 120, a thickness of the
intermediate semiconductor region 120 should be limited to ensure a
sufficiently high number of charge carriers transported between the
first and second terminals 130, 140 will interact with the
interface 180 or the spin selective scattering structure 170.
Hence, a thickness of the intermediate semiconductor region 120
should be below 100 nm, or in different embodiments, below 10 nm,
unless a physical or design process is employed to spatially limit
the extension an effective layer in which the current flows from
the first to the second terminals 130, 140. In the case of a spin
device 100 based on a FET-design, this may be achieved by forming,
during operation, the channel in a spatially limited layer
underneath the interface 180 by applying a voltage to the control
electrode 190.
[0057] Moreover, in embodiments according to the present invention,
the electrical field leading to the forming of the channel may
further attract the charge carriers to the interface 180 to
increase the interaction, leading to an increased
spin-polarization. This biasing effect will be outlined in more
detail below.
[0058] When designing a specific spin device 100, according to an
embodiment of the present invention, a length of the intermediate
semiconductor region 120 and the spin selective scattering
structure 170 may also be considered with respect to the thickness
of the intermediate semiconductor region 120. As will be outlined
in more detail below, a typical length along the direction 210 of
the spin selective scattering structure 170 is in the range of
about 200 nm and above. For instance, when implementing
intermediate semiconductor region 120 in the range of about 1 .mu.m
to several hundred .mu.m (e.g., 500 .mu.m) a technically feasible
degree of spin polarization may be achieved at the second terminal
140. In other words, the length of the spin selective scattering
structure 170 may be in the order of a typical dimension of a
micro-structured device.
[0059] The previously described embodiments according to the
present invention, may also be combined in such a manner that the
control electrode 190 is limited to a central or another smaller
part of the intermediate semiconductor region 120. In this case,
the intermediate semiconductor region 120 may further comprise
doped wells, so that the channel area itself is limited to a
comparably small fraction (e.g., 20 nm to 100 nm) of the
intermediate semiconductor region 120, while the interface 180
extends over the length of more than about 100 nm, or more than
about 1 .mu.m or above.
[0060] Before discussing the possibilities and the general concept
of the working principles of spin devices 100 according to the
present invention in more detail, it should be noted for the sake
of completeness that the spin selective scattering structure 170
typically comprises an insulating material such that the control
electrode 190 is, when positioned above the spin selective
scattering structure 170, electrically insulated from the
intermediate semiconductor region 120 to prevent short-circuits.
Moreover, the spin selective scattering structure 170 may also
comprise additional insulating layers to reduce the risk of a
breakthrough or an avalanche when applying a voltage to the control
electrode 190, which may increase the electrical field beyond a
threshold's characteristic for a spin selective scattering
structure 170. Depending on the implementation in mind, it may for
instance, be advisable to implement a spin device 100 such that the
control voltages of more than 5 V, more than 10 V (e.g., 15 V),
more than 50 V, or other voltages to be applicable to the control
electrode 190. Naturally, a spin device 100 may also be implemented
such that other than the previously mentioned voltage levels may be
used.
[0061] FIG. 1b shows a further embodiment according to the present
invention in the form of a spin device 100, which differs from the
one shown in FIG. 1a mainly with respect to the control electrode
190 and the spin selective scattering structure 170. In the
embodiment of the spin device 100 shown in FIG. 1b, the control
electrode is comprised in the spin selective scattering structure
170, as indicated by a dashed line in FIG. 1b. In other words, the
control electrode 190 is part of the spin selective scattering
structure 170.
[0062] To offer, if advisable, an electric insulation of the
control electrode 190 from the intermediate semiconductor region
120 and the interface 180, the spin selective scattering structure
170 may further comprise an insulating layer arranged between the
control electrode 190 and the interface 180. In other words, in
FIG. 1b, the part of the spin selective scattering structure 170
shown underneath the control electrode 190 may comprise or may be
formed by an insulating layer. Depending on implementation details,
it may be advisable to implement such an insulating layer with
certain boundary conditions. However, this may only be the case for
some implementations, as will be outlined in more details
below.
[0063] Comparing the embodiments shown in FIGS. 1a and 1b, that
further illustrates two different approaches which may be taken in
the case of some embodiments according to the present invention in
the form of a spin device 100. In the case of some, but not all
embodiments according to the present invention, the structures and
layers arranged over the intermediate semiconductor region 120, on
the one hand may be considered to be comprised in the spin
selective scattering structure 170. This may also include the
control electrode 190, for instance, when the control electrode 190
fulfills not only a single function, but also a function with
respect to the spin scattering property of the charge carriers.
This may, for instance, be caused by the material composition of
the control electrode, when, for instance, being fabricated from a
ferromagnetic metal.
[0064] On the other hand, the control electrode 190 may also be
considered not to be part of the spin selective scattering
structure 170, when the spin selective scattering structure is, for
instance, capable of performing its functionality of influencing
the scattering properties of the charge carriers. Both perspectives
may equally well be suitable, when the spin selective scattering
structure 170 is capable of performing its basic functionality
without the control electrode 190, irrespective of the question as
to whether the control electrode may enhance, alter or influence
the spin scattering properties. This may be the case when the
control electrode comprises a ferromagnetic metal.
[0065] A spin device 100 according to an embodiment of the present
invention may offer the opportunity to implement a current source
or a detector for a spin-polarized current on the basis of
semiconductor technology. A spin device 100 may be implemented on
the basis of silicon, as well as other semiconducting materials.
Silicon (Si), for instance, comprises a spin diffusion length over
which the spin polarization decays exponentially, which is large,
when compared to typical dimensions of these devices. Silicon, for
instance, may comprise a spin diffusion length of up to several
centimeters, up to several ten centimeters, so that in the case of
silicon, as a good approximation, the current of the two spin
orientations may be considered independently from one another in
the framework of a so-called two fluid model. Moreover, since in
the case of a FET-like implementation of a spin device 100, the
doping profiles do not necessarily differ from those of FETs, spin
devices 100 based on silicon may be an attractive form of an
implementation of such a device. Since in the case of an
implementation on the basis of an enhancement FET, a voltage is to
be applied to the control electrode 190 to form the inversion
channel at the interface 180, such an implementation may even
provide a comparably high degree of spin polarization, since the
charge carriers will be attracted towards the interface 180 by the
applied voltage, which may increase the interaction with the
selective spin scattering structure 170 (biasing effect).
[0066] However, also other FET structures and other semiconductor
structures may be used in the framework of the intermediate
semiconductor region 120 as the previous discussion has shown. In
principle, even an implementation on the basis of a thick
semiconductor region forming the intermediate semiconductor region
120 ("bulk"-implementation) may be used. However, in such a case
the number of interactions of the charge carriers at the interface
180 might be limited which also leads to a comparably small degree
of spin polarization obtainable at the second terminal 140.
[0067] Spin devices 100 may also be implemented on the basis of
other semiconducting materials, such as materials forming a
two-dimensional electron gas (e.g., AlGaAs). These materials also
offer the possibility of a thin conducting layer close to the
interface 180 between an intermediate semiconductor region 120 and
the spin selective scattering structure 170. However, AlGaAs
typically offer a smaller characteristic spin diffusion length,
which may be below 10 .mu.m, often even below 1 .mu.m.
[0068] FIG. 2a shows a schematic top view of a further spin device
100' according to an embodiment of the present invention. The spin
device 100' is based on a JFET-implementation (JFET=Junction Field
Effect Transistor) with a substrate 110 that comprises an
intermediate semiconductor region 120. T he spin device 100' also
comprises a first terminal 130 and a second terminal 140, which are
adapted to provide current having a first spin polarization to the
intermediate semiconductor region 120 and to extract a current
having a second degree of spin polarization from the intermediate
semiconductor region 120, respectively. As shown in FIG. 1a, the
first terminal 130 may be coupled to a first contact 150 and the
second terminal 140 may be coupled to the second contact 160. A
selective spin selective scattering structure 170, which is shown
as a dashed rectangle in FIG. 2a, is formed on top of the
intermediate semiconductor region 120, to provide an interface 180
between the intermediate semiconductor region 120 and the spin
selective scattering structure 170.
[0069] Moreover, the spin device 100' comprises a first control
electrode 190-1 and a second control electrode 190-2 which may be
coupled to a first control contact 200-1 and a second control
contact 200-2, respectively. However, in contrast to the spin
device 100 shown in FIG. 1a, the two control electrodes 190 are not
electrically insulated from the intermediate semiconductor region
120 by an insulating layer, but by a reverse-biased pn-junction. To
achieve this, the intermediate semiconductor region 120 comprises a
first doping, e.g., an n-type doping, while the control electrodes
190-1, 190-2, comprise a complementary doping, e.g., p-type doping.
As a consequence, when applying a corresponding voltage to the
control electrodes 190, a reverse-biased pn-junction is formed
between the intermediate semiconductor region 120 and the control
electrodes 190. By varying the voltage applied to the control
electrodes 190, a depletion zone can be varied in terms of its
extension so that an electrical field is once again applied
perpendicular to a direction 210 of the current from the first
terminal 130 to the second terminal 140. As a consequence, by
applying the control voltages to the control electrodes 190, the
electrical field perpendicular to the direction 210 influences a
width of the conducting area in the intermediate semiconductor
region 120, which allows control of the magnitude of the
current.
[0070] To describe the structure of the spin device 100' shown in
FIG. 2a in more detail, FIG. 2b shows a cross-sectional view of the
device along a line 250. The spin device 100' is based on a
substrate 110 which comprises an insulating layer 220 which limits
the intermediate semiconductor region 120 in terms of its
thickness. The two control electrodes 190-1, 190-2, are,
accordingly, also limited to the thickness of the intermediate
semiconductor region 120. However, in different implementations of
the spin devices 100', the control electrodes 190-1, 190-2 may
eventually also extend into the insulating layer 220 or may not
penetrate the intermediate semiconductor region 120 at all.
[0071] FIG. 2b together with FIG. 2a furthermore shows that the
spin selective scattering structure 170 does not completely cover
the intermediate semiconductor region 100. Therefore, the interface
180 is not formed over the whole surface of the intermediate
semiconductor region 120. This implementational detail is, however,
by far not limited to spin devices 100' in the form of JFET-like
devices. In principle, the spin selective scattering structure 170
may also be limited in terms of its extension in the case of
FET-like devices, as shown in FIG. 1a.
[0072] Before describing a third embodiment according to the
present invention with respect to FIG. 3, it should be noted that
structures and objects with identical or similar functional or
structural properties will be denoted with the same or similar
reference signs. For example, the spin device 100 shown in FIG. 1a
and the spin device 100' are denoted by similar reference signs.
Moreover, unless noted otherwise, structures and objects denoted by
the same or similar reference signs may be implemented equally
having equal electrical, mechanical, or other properties and
dimensions. Naturally, objects and structures denoted by the same
or similar reference signs may be implemented differently.
[0073] Moreover, summarizing reference signs will be used for
objects and structures appearing more than once in an embodiment or
a figure. For example, in FIGS. 2a and 2b, the two control
electrodes 190-1, 190-2 are referred to by their summarizing
reference signs 190, unless specifically referring to a control
electrode. The spin devices 100, 100' shown in FIGS. 1 and 2,
respectively, will be referred to by their summarizing reference
sign 100, unless a specific detail of either of the spin devices is
referred to.
[0074] FIG. 3 shows a further embodiment according to the present
invention of a spin device 100'', which is similar to the spin
device 100 shown in FIG. 1a. The spin device 100'' of FIG. 3 also
comprises a substrate 110, which in turn comprises an intermediate
semiconductor region 120. The intermediate semiconductor region 120
is arranged between a first terminal 130 and a second terminal 140,
which are contacted by a first contact 150 and a second contact
160, respectively.
[0075] The spin device 100'' further comprises a selective spin
scattering structure 170 abutting the intermediate semiconductor
region 120 to form an interface 180 between the spin selective
scattering structure 170 and the intermediate conductor region 120.
However, in contrast to the embodiment shown in FIG. 1a, the spin
selective scattering structure 170 does not extend over the whole
length of the intermediate semiconductor region 120, as shown in
the cross-sectional view of FIG. 3. The spin selective scattering
structure 170 only covers a fraction of a surface of the
intermediate semiconductor region 120 such that this fraction of
the surface forms the interface 180 only, by which charge carriers
injected into the intermediate semiconductor region 120 from the
first terminal 130 interact to alter the degree of spin
polarization of the current.
[0076] The spin device 100'' shown in FIG. 3 also comprises a
control electrode 190, which is electrically connectable to a
further circuit or element by a control contact 200. However, the
control electrode 190 is electrically insulated from the
intermediate semiconductor region 120 by a further insulating layer
270 abutting the intermediate semiconductor region 120. The further
insulating layer 270 and the intermediate semiconductor region 120,
hence, form a further interface 280, which may abut the interface
180, but is also separated from this.
[0077] The spin device 100'' shown in FIG. 3, therefore, differs
from the spin device 100 shown in FIG. 1a by the control electrode
190 being laterally displaced with respect to the spin selective
scattering structure 170. While in the embodiment shown in FIG. 1a
the control electrode 190 is arranged on top of the spin selective
scattering structure 170, which also electrically insulates the
control electrode from the intermediate semiconductor region 120,
the electrical insulation is realized in the embodiment shown in
FIG. 3 by the further insulating layer 270. As a result, the
functional parts of spin selectively scattering the charge carriers
by the spin selective scattering structure 170 and the respective
interface 180 on the one hand, and controlling the magnitude of the
current by applying an electrical field perpendicular to a
direction 210 of the current between the first terminal 130 and the
second terminal 140 on the other hand, may spatially be separated
from one another. This may enable, for instance, separately
optimizing the spin selective scattering structure 170 on the one
hand and the controlling structure controlling the magnitude of the
current comprising, in the embodiment shown in FIG. 3, the further
insulating layer 270, the control electrode 190 and the control
contact 200 on the other hand.
[0078] It may happen, for instance, that the spin selective
scattering structure 170 is influenced by the electrical field
applied by the control electrode 190. This influence may be
reduced, or even omitted, by spatially separating both. Moreover,
by introducing the further insulating layer 270, the control
electrode 190 may be operated at different voltage levels, since a
thickness of the further insulating layer 270 may be chosen
independently of the spin selective scattering structure 170. It
may, therefore, be possible to operate the spin device 100'' with
higher control voltages applied to the control electrode 190, due
to the possibility of implementing a thicker further insulating
layer 270 without surpassing a breakthrough electrical field of the
further insulating layer 270, which may cause an avalanche of
charge carriers and the creation of a conducting channel through
the further insulating layer 270 destroying the spin device
100''.
[0079] In other words, the spin selective scattering structure 170
and the controlling structure comprising the control electrode 190
form a series connection with the spin selective scattering
structure 170 being coupled before the controlling structure, with
respect to the direction of the current 210. In different
embodiments, the order of the two structures may equally well be
reversed, placing the spin selective scattering structure 170
behind the controlling structure.
[0080] However, as a possible drawback of this solution, the
surface of the intermediate semiconductor region 120 may eventually
not be used as efficiently as in the case of the spin device 100
shown in FIG. 1a, since the surface of the intermediate
semiconductor region 120 in FIG. 3 comprises two distinct areas
forming the interface 180 and the further interface 280. This may,
eventually, not be a significant drawback, since depending on
implementation details, the further interface 280 may eventually be
significantly smaller in terms of its length along the direction of
the current 210, compared to a length of the interface 180.
[0081] In the case of a FET-like implementation of the spin device
100'', the channel or channel area will mostly be formed in the
area of the further interface 280. Hence, the length of the further
interface 280 may, in principle, be limited to a minimal length
necessary to control the admissible current. Depending on
implementation details, the length of the further interface 280
may, in principle, be limited to less than about 100 nm, while the
length of the interface 180 is longer than about 1 .mu.m.
[0082] As an optional implementation, the spin device 100'' may
also comprise an insulating layer 220, as already shown and
discussed in the framework of the spin device 100 shown in FIG. 1a.
As all insulating layers described herein, the insulating layer
220, as well as the further insulation layer 270 may comprise an
oxide material, a nitrite material, an oxynitride material, or
another insulating material (e.g., ONO (Oxide-Nitride-Oxide)).
[0083] Furthermore, although FIG. 3 shows the spin selective
scattering structure 170 being positioned before the control
electrode 190 along with the further insulating layer 270, the
direction of the current 210 may equally well be reversed, as
hinted to before. In principle, the spin device 100'' may be
designed symmetrically so that the second terminal 140 shown in
FIG. 3 may also be considered to be the first terminal 130 and vice
versa. Accordingly, by inverting the direction 210 of the current,
the control electrode 190 may likewise be arranged before the spin
selective scattering structure 170. With respect to the operational
principle, this is of no importance. In yet other words, an order
of the arrangement of the control electrode 190 and the spin
selective scattering structure 170 may be reversed as shown in FIG.
3.
[0084] So far, the description of the embodiments as shown in FIGS.
1 to 3 have mainly focused on the arrangement of the spin selective
scattering structure 170 with respect to the control electrodes 190
and the different principles on which a spin device 100 may be
implemented. In the following, a possible implementation of a spin
selective scattering structure 170 will be described in more detail
in context with FIGS. 4a and 4b.
[0085] Based on fundamental finding, embodiments according to the
present invention offer a new method for a controllable selective
spin injection, for example, by applying a gate or control
voltage.
[0086] FIG. 4a shows an arrangement 300 for evaluating the
possibility of spin-dependence scattering of electrons at an
interface between a semiconductor and a thin, wedge-shaped
magnesium oxide layer (MgO) covered by a ferromagnetic material.
The arrangement 300 comprises a semiconductor region 310 and a
wedge-shaped oxide layer 320 based on MgO so that an interface 330
is formed between the semiconductor region 310 and the oxide layer
320. In the arrangement 300 shown in FIG. 4a, a thickness of the
oxide layer 320 rises from 0.2 nm (2 .ANG.) and a thickness of more
than 1.5 nm (15 .ANG.). A ferromagnetic layer 340 is positioned on
top of the oxide layer 320.
[0087] The ferromagnetic layer 340 comprises a sufficient thickness
and a sufficient area to enable the ferromagnetic layer 340 to form
domains such that the ferromagnetic layer 340 may comprise a (more
or less constant) magnetization 350. The ferromagnetic layer 340
may be formed by any ferromagnetic material such as ferromagnetic
metals, iron (Fe), cobalt (Co), and nickel (Ni). Moreover, as a
ferromagnetic material more complex compounds may be used such as
rare earth magnets comprising alloys of rare earth elements (e.g.,
neodymium magnets (Nd.sub.2Fe.sub.14B), samarium magnets
(SmCo.sub.5)). Ferrite magnets with a chemical formula
AB.sub.2O.sub.4 wherein A and B represent various metal cations,
usually including iron (e.g., Zinc Ferrite "ZnFe",
ZnFe.sub.2O.sub.4), as well as "Alnico" magnets comprising an alloy
of aluminum (Al), nickel (Ni), cobalt (Co), along with optional
additional iron (Fe), copper (Cu), and sometimes titanium (Ti),
Heusler alloys, magnetite (Fe.sub.3O.sub.4), doped insulators and
semiconductors (e.g., doped with manganese (Mn)) and other
conventionally known ferromagnetic materials (e.g.,
FeOFe.sub.2O.sub.3, NiOFe.sub.2O.sub.3, CuO Fe2O3, MgO
Fe.sub.2O.sub.3, CrO.sub.2, and EuO may also be used.
[0088] The following experimental results, are however, based on
iron (Fe) as the ferromagnetic material of the ferromagnetic layer
340 and magnesium oxide (MgO) for the oxide layer 320. Depending on
the thickness of the oxide layer 320 and the orientation or the
magnetization 350 of the ferromagnetic layer 340, different kinds
of spin orientated electrons may be obtainable after a reflection
at or an interaction with the interface 330.
[0089] Starting at comparably small thicknesses of the oxide layer
320 of less than 0.2 nm (2.0 .ANG.), electrons are preferably
scattered at the interface 330 with a spin direction parallel to
the magnetization 350 of the ferromagnetic layer 340, as
schematically indicated in FIG. 4a, an electron 360 having either
spin polarization before the scattering event and being reflected
with a preferable spin orientation to the magnetization 350. This
spin selective scattering property which is also sometimes referred
to as ferromagnetic proximity polarization (FPP) is followed by a
steep reduction with increasing thicknesses of the oxide layer 320
and a sign reversal, as schematically indicated by an electron 370
in FIG. 4a.
[0090] As FIG. 4a schematically shows, the electron 370 approaches
the interface 330 with either spin polarization and will be
scattered preferably with a spin orientation opposite to the
magnetization 350 of the ferromagnetic layer 340. While the FPP
spin polarization is enhanced with a thickness of approximately 2.0
.ANG. (0.2 nm) of magnesium oxide for the schematically shown
electron 360, the spin polarization comprises the opposite sign
when the magnesium oxide thickness is approximately 7.0 (0.70 nm)
to 7.9 .ANG. (0.79 nm). At a thickness of approximately 5.0 .ANG.
(0.5 nm), the FPP spin polarization crosses zero so that the spin
polarizing effect is significantly suppressed.
[0091] Increasing the thickness of the oxide layer 320 over the
previously mentioned 7.0 to 7.9 .ANG., at which the sign reversal
of the FPP spin polarization occurs will once again results in a
reduction and, finally, into a vanishing of the FPP at
approximately 12 .ANG. (0.12 nm) to 14.0 .ANG. (0.14 nm), 15.0
.ANG. (0.15 nm) and above. This is schematically illustrated by an
electron 380, which after a scattering process at the interface
330, does not comprise a preferred spin orientation.
[0092] The experiment described above has been carried out based on
the longitudinal Magneto-Optic Kerr Effect (MOKE) on the basis of
magnesium oxide layers and a Gallium Arsenide layer (GaAs) on an
Aluminum-Gallium-Arsenide/Gallium Arsenide substrate
(Al.sub.0.7Ga.sub.0.3As/GaAs). The thickness of the ferromagnetic
layer 340 was during the experiments 2.0 nm (20 .ANG.). However, as
illustrated before, the ferromagnetic layer 340 may comprise
thicknesses of 1 nm (10 .ANG.) and above, for instance, 13 .ANG.
(1.3 nm) or other thicknesses depending on different
parameters.
[0093] A number of explanations and theories exist as to the reason
for the observed behavior. Among others, an interfacial bonding,
magnesium oxide barrier height effects, quantum interference or
surface effects in or at the magnesium oxide layer (due to
non-perfect surfaces), localized extended states and majority-spin
bulk states have been proposed as explanations for the observed
behavior. Moreover, spin-orbit-coupling has also been proposed as
an explanation for the observed spin dependency of the reflected
electrons at the surface 330.
[0094] The experimental observations have shown so far effects of
about 30% of the resulting spin-polarized current. However,
depending on the experimental circumstances and other technological
parameters, higher or lower spin polarization degrees may be
obtainable.
[0095] Such a simple structure with a magnesium oxide wedge 320
covering the interface 330 between the semiconductor 310 and the
thin magnesium oxide layer 320, which in turn is covered by the
ferromagnetic layer 340, does not allow specification of the
resulting orientation of the scattered electron spins, since
electrons may interact with the interface 330 at locations
corresponding to different thicknesses of the oxide. For example,
along a first path 390-1, as shown in FIG. 4b, scattered electrons
will comprise a different spin orientation with a higher
probability than along a second path 390-2, also shown in FIG. 4b.
Therefore, the experimental setup as shown in FIGS. 4a and 4b,
which only differ with respect to the additionally marked paths 390
from FIG. 4a, is not very suitable for an application in the
framework of a spin device. FIG. 4b illustrates well that the spin
orientation depends on the magnesium oxide thickness relevant for
the shown individual electron paths 390-1, 390-2.
[0096] In contrast to the experimental setup shown in FIGS. 4a and
4b, using a uniform magnesium oxide layer with a constant
thickness, offers the possibility of obtaining a fixed spin
scattering property depending on the magnesium oxide thickness and
the relevant magnetization of the ferromagnetic layer 340.
[0097] It may eventually be possible to alter the spin scattering
property by applying a voltage perpendicular to the interface 330
of a stack 300, since by applying such a voltage to an electrode
(not shown in FIG. 4a and 4b) may enable varying strengths of a
spin-orbit coupling leading to a virtual change of the "thickness"
of the insulating layer 320. Therefore, it may be possible to
obtain, with a single spin selective scattering structure based on
the arrangement 300 alone, two different spin polarizations, simply
by adjusting a control voltage to create an electrical field
perpendicular to the interface 330. In other words, it may be
possible that by adjusting the control voltages supplied to the
control electrode 190 of the spin device 100 shown in FIG. 1a, in
case the spin selective scattering structure 170 comprises an
arrangement similar to the arrangement 300 with an oxide layer 320
comprising a constant thickness and an additional insulating layer
insulating the ferromagnetic material from the control electrode
190 above, to switch between different orientations or different
degrees of spin polarization by simply adjusting the control
voltage. In yet other words, it may be possible to switch between
the two paths 390-1, 390-2 by adjusting the respective control
voltage.
[0098] In a technical implementation of a spin selective scattering
structure 170, the oxide layer 320 is typically deposited having a
constant thickness. Typical values for the thickness in the case of
a magnesium oxide layer vary between approximately 2 and 10 atomic
layers, sometimes between 5 and 10 atomic layers. This will result
in thicknesses between approximately 2 and 8 .ANG., while for
thicknesses above 12 .ANG. the effect appears to be vanishing.
[0099] In implementations of a spin device 100 according to an
embodiment of the present invention, also other insulating layers
instead of the oxide layer 320 comprising magnesium oxide may
eventually be used. Magnesium oxide offers the possibility of a
highly reproducible selective spin scattering structure and the
technological process of employing magnesium oxide is well founded
since magnesium oxide is, for instance, used in TMR-related
fabrication processes (TMR=Tunneling Magneto-Resistance). Other
materials, which may be used instead of magnesium oxide, comprise,
for instance, aluminum oxide (AlO.sub.x) which is frequently also
used as a tunneling barrier in the case of TMR-based systems. As a
further alternative, the ferromagnetic layer 340 and the oxide
layer 320 may eventually be replaced by a single compound such as
magnetite (Fe.sub.3O.sub.4) which is a ferromagnetic insulator. In
other words, the spin selective scattering structure 170 as shown
in FIGS. 1 to 3 may eventually be formed by a sufficiently thick
magnetite layer.
[0100] In the following figures, several embodiments according to
the present invention in the form of a spin device 100 will be
shown in more detail compared to the schematic representations of
the FIGS. 1 to 3. In the following figures, the spin selective
scattering structure 170, which was only schematically shown in
FIGS. 1 to 3, will be described in more detail. However, it should
be pointed out again that the embodiments shown in FIGS. 1 to 3 may
easily be implemented by implementing the spin selective scattering
structure comprising a magnetite layer or another suitable
structure.
[0101] Optionally, the spin selective scattering structure may
comprise an additional insulating layer, for instance, a magnesium
oxide layer forming the interface 180 underneath the magnetite or
it may further comprise an insulating layer on top of the magnetite
layer to enhance the electrical insulation to the control electrode
190 deposited on top of the spin selective scattering structure
170, in the case of a spin device 100 as shown in FIG. 1a.
[0102] FIG. 5a shows a cross-sectional view of a spin device 100
according to an embodiment of the present invention based on a
FET-structure which is fairly similar to the spin device 100 shown
in FIG. 1a. The spin device 100 of FIG. 5a also comprises a
substrate 110 comprising an intermediate semiconductor region 120
in which a channel will be formed during operation. Accordingly, a
first terminal 130 abutting the intermediate semiconductor region
120 is also referred to as a source contact. A second terminal 140,
also abutting the intermediate semiconductor region 120, is
therefore also referred to as a drain terminal.
[0103] The intermediate semiconductor region 120 abuts the spin
selective scattering structure 170 so that an interface 180 is
formed between the two. On top of the spin selective scattering
structure 170, a control electrode or a gate electrode is
deposited, which is electrically connected to a control contact 200
which is also referred to as the gate contact.
[0104] However, the spin device 100 of FIG. 5a differs from the
spin device 100 of FIG. 1a with respect to the spin selective
scattering structure 170. To be more precise, the spin selective
scattering structure 170 of FIG. 5a comprises a magnesium oxide
layer 400 (with a constant thickness as compared to the wedge-shape
of layer 320 in FIGS. 4a and 4b) abutting the intermediate
semiconductor region 120 and forming the interface 180. On top of
the magnesium oxide layer 400, a ferromagnetic layer 410 is
deposited. The ferromagnetic layer 410 along with the magnesium
oxide layer 400 form the spin selective scattering structure 170.
As laid out before, the ferromagnetic layer 410 may be an
insulating layer (e.g., magnetite), a semiconducting layer (e.g.,
magnetically doped semiconductor) or a conductive or metallic layer
(e.g., iron (Fe), nickel (Ni) or cobalt (Co)).
[0105] It should be noted that different materials may be used in
different embodiments according to the present invention. For
instance, the magnesium oxide layer 400 may be replaced by another
electrically insulating layer. This layer may, for instance,
comprise aluminum oxide (AlO.sub.x), silicon dioxide (SiO.sub.2),
silicon oxide (SiO), a nitride (e.g., Si.sub.xN.sub.y (with x, y
indicating the stoichiometrical composition, e.g., Si.sub.3N.sub.4)
or a combination thereof In other words, in the embodiment shown in
FIG. 5a, the magnesium oxide layer 400, a as a generalization, may
be replaced by another insulating layer, as previously described,
to obtain a spin device 100 according to a different embodiment of
the present invention.
[0106] It might be advisable to electrically insulate the
ferromagnetic layer 410 from the gate control 190 via an insulation
layer in order to prevent influences on the ferromagnetic
properties by an electrical current going into the ferromagnetic
layer 410, if fabricated from a ferromagnetic metal or another
ferromagnetic conductor. For instance, in the case of using iron
(Fe) as the ferromagnetic layer 410, this may be advisable, but is
by far not necessary. The spin device 100 comprising such an
electrical insulation is shown in FIG. 5b.
[0107] FIG. 5b shows a further spin device 100 according to an
embodiment of the present invention, which differs from the spin
device 100 shown in FIG. 5a merely with respect to an additional
gate oxide layer 420, which is arranged in between the
ferromagnetic layer 410 and the control electrode 190.
[0108] The gate oxide layer 420 in between the ferromagnetic layer
410 and the gate electrode or control electrode 190 offers the
possibility of adjusting the effective electrical insulation of the
gate electrode 190 more independently from the electrical
properties of the spin selective scattering structure 170
comprising the stack of the magnesium oxide layer 400 and the
ferromagnetic layer 410.
[0109] This may eventually be advisable, since a thickness of the
magnesium oxide layer 400, as well as a thickness of the
ferromagnetic layer 410 may eventually be defined due to the
functional properties as being part of the spin selective
scattering structure 170. As a consequence, choosing the control or
gate voltage to be applied to the gate electrode 190 may be
limited. For instance, applying values of more than 10 V (e.g., 15
V) to the control electrode 190 may not be possible without
introducing the gate oxide layer 420. In other words, by
introducing the gate oxide layer 420 and, therefore, by replacing
the stack of the spin selective scattering structure 170 as shown
in FIG. 5a comprising the ferromagnetic layer 410 and the magnesium
oxide layer 400 by the more complicated stack comprising also the
gate oxide layer 420 may eventually lead to more flexible and to a
more applicable spin device 100.
[0110] Depending on the perspective, the gate oxide layer 420 may
well be considered to be part of the spin selective scattering
structure 170. In this case, the spin selective scattering
structure 170 comprises the stack of the three layers 400, 410 and
420 as previously outlined. In a further embodiment according to
the present invention, based on the embodiment shown in FIG. 5b, a
further gate oxide may eventually be introduced in between the
magnesium oxide layer 400 and the ferromagnetic layer 410, which
may eventually be advisable to strengthen the voltage resistance
with respect to electrical fields applied via the control or gate
electrode 190. However, depending on the concrete implementation
and the circumstances, it may eventually be advisable to implement
the gate oxide layer in the form of a magnetite layer (magnetite
also being an oxide of iron). Typical thicknesses of the gate oxide
layer 420 and of a further gate oxide layer may be chosen from
typical values in the field of semiconductor processing, bearing
the planned operational parameters in mind.
[0111] As a further application, the gate oxide layer 420 may be
used as a tunneling barrier, when the ferromagnetic layer 410 is a
conductive or metallic layer. In this case, the ferromagnetic layer
410 may be used as a floating gate electrode comparable to a flash
memory device. In this case, by tunneling charge carriers from the
control electrode 190 onto the ferromagnetic layer 410, a charge
state of the ferromagnetic layer 410 may be changed such that the
altered charge state creates an additional electrical field
perpendicular to a direction 210 of the current inside the
intermediate semiconductor region 120. This additional electrical
field may then contribute to the electrical field perpendicular to
the direction 210 of the current controlling the magnitude of the
current. Hence, the spin device may be considered to be
programmable by introducing a floating gate-like layer, as will be
outlined in more detail below.
[0112] FIG. 6a shows a further embodiment according to the present
invention of a spin device 100. The spin device 100 is based on a
FET-like structure comprising a substrate 110 with an intermediate
semiconductor region 120 abutting a source terminal 130 (first
terminal) and a drain terminal 140 (second terminal). The spin
device 100 further comprises a spin selective scattering structure
170 comprising a magnesium oxide layer 400 and a ferromagnetic
layer 410 as shown in FIGS. 5a and 5b. The spin selective
scattering structure 170 forms an interface 180 with the
intermediate semiconductor region 120.
[0113] The spin device 100, as shown in FIG. 6a, comprises a
floating gate electrode 430 on top of the ferromagnetic layer 410,
which may, for instance, be fabricated from poly-silicon. The
insulating layer 440 on top of the floating gate electrode 430, may
for instance be formed by an oxide material or an
Oxide-Nitride-Oxide layer (ONO). On top of the insulating layer
440, a control electrode or a gate electrode 190 is deposited along
with a control contact 200.
[0114] By introducing the floating gate electrode 430, which is
electrically insulated from the control or gate electrode 190 via
the insulating layer 440, charge carriers may tunnel from the
control electrode 190 onto the floating gate electrode 430, which
may lead to a change of the charge state of the floating gate
electrode 430. In other words, the floating gate electrode 430 may
be charged or discharged via the control electrode 190 by means of
tunneling through the insulating layer 440.
[0115] By changing the charge state of the floating gate electrode
430, electrostatic charges are provided to the floating gate
electrode 430 which may contribute to the electrical field
perpendicular to the direction 210 of the current between the first
terminal 130 and the second terminal 140. As a result, it may be
possible to "program" the spin device 100 by pre-charging the
floating gate electrode 430 accordingly. In other words, by
providing the floating gate electrode 430 with an appropriate
charge, the spin device 100 may be programmed such that in the case
of an implementation based on an enhancement FET, the spin device
100 may be turned on even without a control voltage to the control
electrode 190. Naturally, by applying additional voltage to the
control electrode 190 (either positive or negative), the magnitude
of the current flowing between the first, and the second terminals
130, 140, is still controllable.
[0116] It should be noted that the ferromagnetic layer 410 may once
again be an electrically conducting or an insulating layer 410. In
the case of an electrically conducting ferromagnetic layer 410
(e.g., a ferromagnetic metal), the floating gate electrode 430
together with the ferromagnetic layer 410 acts as the "floating
gate electrode".
[0117] Moreover, depending on the perspective as outlined in the
context of FIG. 5b, the spin selective scattering structure 170 may
also be considered to comprise the floating gate electrode 430, as
well as the insulating layer 440.
[0118] FIG. 6b shows a further embodiment according to the present
invention of a spin device 100, which is similar to the spin device
100 shown in FIG. 6a. The spin device 100 shown in FIG. 6b differs
from the spin device 100 shown in FIG. 6a mainly with respect to an
additional insulating layer 450 which is arranged between the
ferromagnetic layer 410 and the floating gate electrode 430.
[0119] By introducing a further insulating layer 450, the
ferromagnetic layer 410 is electrically insulated from the floating
gate electrode 430. As a result, in the case of an electrically
conducting ferromagnetic layer 410 (e.g., ferromagnetic metal)
charging of the ferromagnetic layer 410 may be prevented, which may
eventually lead to a disturbance of the electrical and/or magnetic
properties of the layer 410.
[0120] The further insulating layer 450 may, for instance, comprise
an oxide material, which may for instance be used in other
semiconductor-related fabrication processes and devices as a tunnel
oxide. However, other insulating materials may also be used
here.
[0121] FIG. 7a shows a spin device 100 according to a further
embodiment of the present invention, which is similar to the
embodiment shown in FIG. 3. Once again, the spin device 100
comprises laterally and geometrically decoupled components or
structures for the spin selective scattering on the one hand, and
controlling the magnitude of the current on the other hand. To
achieve this, the intermediate semiconductor region 120 comprised
in the substrate 110 forms an interface 180 with respect to the
spin selective scattering structure 170 and a further interface 280
with respect to a further insulating layer 270, the control
electrode 190, and the control contact 200 on the other hand. The
surface of the intermediate semiconductor region 120, hence,
comprises at least the interface 180 used during the spin selective
scattering and the further interface 280 in the vicinity of which
the control of the magnitude of the flowing current is
performed.
[0122] The spin selective scattering structure 170 comprises a
magnesium oxide layer 400 and a ferromagnetic layer 410, as shown
in FIG. 5a. The magnesium oxide layer 400 abuts the intermediate
semiconductor region 120 at least partially, so that the interface
180 is formed between the two.
[0123] The spin device 100 as shown in FIG. 7a is again a spin
device based on a FET-design. As a consequence, the first terminal
130 is also referred to as a source terminal, whereas the second
terminal 140 is also referred to as the drain terminal. As a
result, the further insulating layer 270 is also referred to as a
gate oxide (GOX), while the control electrode 190 is referred to as
the gate electrode. The control contact 200 is, therefore, also
referred to as the gate contact.
[0124] In the spin device 100 as shown in FIG. 7a, the current
first passes the spin selective scattering structure 170 and is
afterwards controlled with respect to its magnitude by the
electrical field perpendicular to the direction of the current 210
by applying the corresponding gate voltage to the control electrode
190. The spin selective scattering structure 170 and the control
electrode 190 form together with the further insulating layer 270 a
"series connection", wherein the spin selective scattering
structure 170 is coupled before the control electrode 190.
[0125] However, the order of the spin selective scattering
structure 170 and the control electrode 190 is by far not
mandatory. As outlined in the context of FIG. 3, the flow of the
current may be equally well reversed.
[0126] FIG. 7b shows a corresponding spin device 100 which differs
from the spin device 100 shown in FIG. 7a mainly with respect to
the order of the spin selective scattering structure 170 and the
control electrode 190. In other words, in the case of the spin
device 100 as shown in FIG. 7b, the spin selective scattering
structure 170 is coupled behind the control electrode 190.
[0127] It should be noted that the spin devices 100 according to
embodiments of the present invention as shown in FIGS. 7a and 7b
may lack the possibility of influencing the spin orientation by
applying an appropriate voltage to the spin selective scattering
structure, but may therefore offer the possibility of decoupling
the spin selective scattering structure 170 from the influences of
the vertically applied electrical field of the control electrode
190. It may therefore be possible to reduce or even omit unwanted
parasitic effects caused by the perpendicular electrical field.
[0128] FIG. 8a shows a further embodiment according to the present
invention, which is similar to the spin device 100 as shown in FIG.
7b. To be more precise, the spin device 100 as shown in FIG. 8a
differs only with respect to the spin selective scattering
structure 170. Apart from the magnesium oxide layer 400 forming the
interface 180 along with at least part of the intermediate
semiconductor region 120 and the ferromagnetic layer 410, the spin
selective scattering structure 170 further comprise a biasing
contact 460. Therefore, the ferromagnetic layer 410 may be used as
a biasing electrode 470, in the case of using a ferromagnetic
conductive material (e.g., ferromagnetic metal), as the
ferromagnetic layer 410.
[0129] By introducing such a biasing contact 460, the spin
selective scattering properties of the spin selective scattering
structure 170 may eventually be altered. By applying the
corresponding voltage, the charge carriers of the current may be
attracted towards the interface 180 so that the spin selective
scattering properties may be enhanced by increasing the number of
interactions of the charge carriers with the interface 180. As a
result, it may be possible to influence the degree of spin
polarization obtainable.
[0130] FIG. 8b shows a cross-sectional view of a further spin
device 100 according to an embodiment of the present invention. The
spin device 100 of FIG. 8b differs from the spin device 100 shown
in FIG. 7a mainly with respect to an additional biasing electrode
470, which is electrically coupled to a biasing contact 460. The
biasing electrode 470 is arranged on top of the ferromagnetic layer
410 and separated from it by an insulating layer 480 covering the
ferromagnetic layer 410 at least in an area in which the biasing
electrode 470 is deposited. The insulating layer 480, hence,
electrically insulates the biasing electrode 470 from the
ferromagnetic layer 410.
[0131] By electrically insulating the biasing electrode 470 from
the ferromagnetic layer 410, it may be possible to reduce unwanted
influences, which may be caused by a current flowing into the
ferromagnetic layer 410 due to a direct contact of the
ferromagnetic layer 410 with the biasing contact 460 as it is in
the case of the spin device 100 shown in FIG. 8a. The spin device
100 of FIG. 8b may offer the possibility of applying a voltage to
the biasing electrode 470 in such a way that the charge carriers of
the current flowing from the first terminal 130 (source terminal)
to the second terminal 140 (drain terminal) may be attracted
towards the interface 180. This may lead to increasing the number
of interactions of the charge carriers with the spin selective
scattering structure 170 comprising not only the magnesium oxide
layer 400, the ferromagnetic layer 410, but also the biasing
electrode 470, the insulating layer 480 and the biasing contact
460. Hence, compared to the spin device as shown in FIG. 7a,
implementing the biasing electrode 470 may once again offer the
possibility of enhancing the spin polarizing defect of the spin
device 100.
[0132] It should be noted for the sake of completeness that once
again the direction of the current is, by far, not limited to
current flowing from the first terminal 130 to the second terminal
140. The order of the spin selective scattering structure 170 and
the control structure comprising the further insulating layer 270
(e.g., gate oxide) along with the control electrode 190 (gate
electrode) may be reversed, not only for the spin device 100 shown
in FIG. 8a, but also for the spin device 100 shown in FIG. 8b. In
other words, the control electrode 190 may either be arranged
before or behind the spin selective scattering structure 170.
[0133] FIG. 9a shows a further spin device 100 according to the
embodiment of the present invention with spatially and
geometrically separated components concerning the spin scattering
and the controlling of the magnitude of the current flowing between
the first terminal 130 and the second terminal 140 of the spin
device 100. The spin device 100 of FIG. 9a comprises a similar
structure compared to the spin device 100 shown in FIG. 7b. It
mainly differs from its counterpart in FIG. 7b with respect to the
component for controlling the magnitude of the current flowing from
the first terminal 130 to the second terminal 140. While the spin
selective scattering structure 170 of the spin device 100 of FIG.
9a is identical to the spin selective scattering structure 170 of
FIG. 7b, the spin device 100 of FIG. 9a comprises a floating gate
electrode 430 and an insulating layer 440 (e.g., an oxide or ONO
(Oxide-Nitride-Oxide)). As a consequence, the floating gate
electrode 430, which may be fabricated from poly-silicon, is
electrically insulated from the intermediate semiconductor region
120 by the further insulating layer 270 and from the control
electrode 190 (control gate) by the insulating layer 440.
[0134] As already outlined in the context of FIGS. 6a and 6b, this
may offer the possibility of providing an electrical charge to the
floating gate electrode 430 to change its charge state, so that the
charge, as on the floating gate electrode 430, generates an
electrical field perpendicular to the direction 210 of the current
from the first terminal 130 to the second terminal 140. This
electrical field is superimposed to the perpendicular electrical
field caused by charging the control electrode 190 during operation
of the spin device 100. As a consequence, by charging the floating
gate electrode 430, a default state of the spin device 100 may be
influenced, for instance, to increase or decrease a special voltage
to be applied to the control electrode 190 in order to switch on or
switch off the current in the intermediate semiconductor region
120.
[0135] The floating gate electrode 430 may be charged by tunneling
charge carriers from the control electrode 190 to the floating gate
electrode 430 passing the insulating layer 440 or from the
intermediate semiconductor region 120 passing the further
insulating layer 270. Depending on the concrete implementation of
the spin device 100, the further insulating layer 270 may, for
instance, may also be referred to as a tunnel oxide, when the
charging or discharging of the floating gate electrode 430 is
accomplished via the intermediate semiconductor region 120.
Applying a respective voltage to the intermediate semiconductor
region 120 may, for instance, be accomplished by providing the
corresponding voltage to any or to both of the terminals 130,
140.
[0136] On the other hand, the charge carriers may also tunnel onto
the floating gate electrode 430 by passing the insulating layer 440
when an appropriate voltage is applied to the control electrode
190.
[0137] FIG. 9b shows a further embodiment according to the present
invention in the form of a spin device 100, which differs from the
spin device 100 of FIG. 9a mainly with respect to the order, in
which the spin selective scattering structure 170 and the structure
for controlling the magnitude of the current comprising the further
insulating layer 270, the floating gate electrode 430, the
insulating layer 440 and the control electrode 190 are arranged. In
the case of the spin device 100 of FIG. 9b the structure for
controlling the magnitude of the current is arranged behind the
spin selective scattering structure 170 with respect to the
direction 210 of the current flowing from the first terminal 130 to
the second terminal 140.
[0138] FIG. 10a shows a further embodiment according to the present
invention in the form of the spin device 100, which is similar to
the spin device 100 of FIG. 9a. The spin device 100 of FIG. 10a
differs from the spin device 100 of FIG. 9a with respect to the
spin selective structure 170 comprising a biasing contact 460,
which allows providing a biasing voltage to the ferromagnetic layer
410 to increase the number of interactions of charge carriers in
the intermediate semiconductor region 120 with the interface 180.
As outlined in the context of FIGS. 8a and 8b, this may result in
an increased number of interactions since the charge carriers are
attracted towards the interface 180 so that the degree of spin
polarized charge carriers of the current may be increased compared
to the spin device 100 shown in FIG. 9a. As a consequence, the
ferromagnetic layer 410 may be used as a biasing electrode 470 in
the case that the ferromagnetic layer 410 is a conducting layer
(e.g., a metallic layer).
[0139] Apart from that, the spin device 100 also comprises the
floating gate electrode 430, which may be charged via the
insulating layer 440 electrically insulating the control electrode
190 from the floating gate electrode 430 or via the further
insulating layer 270 (e.g., tunnel oxide) insulating the floating
gate electrode 430 from the intermediate semiconductor region
120.
[0140] FIG. 10b shows a further spin device 100 according to an
embodiment of the present invention, which is similar to the spin
device 100 of FIG. 9b with respect to the structure comprising the
control electrode 190, the insulating layer 440, the floating gate
electrode 430 and the further insulating layer 270 as well as the
reversed current direction compared to the spin device 100 shown in
FIG. 10a. However, with respect to the spin selective scattering
structure 170, the spin device 100 further comprises a separately
implemented biasing electrode 470, which is electrically insulated
from the ferromagnetic layer 410 by an insulating layer 480
deposited in between the two previously mentioned layers. The
biasing electrode 470 may be then electrically coupled via a
biasing contact 460 to further circuits or structures.
[0141] As outlined before in the context of FIG. 8b, the insulating
layer 480 is not required to cover the whole ferromagnetic layer
410 with only the region over which the biasing electrode 470 is
deposited. Apart from preventing unwanted influences on the
ferromagnetic layer 410, using a separate biasing electrode 470 may
also be advisable in case the ferromagnetic layer is an insulating
layer (e.g., magnetite). Eventually, the insulating layer 480 may
be omitted in such a case.
[0142] FIGS. 11a and 11b show a top view of a spin device 100
implemented as a lateral device according to an embodiment of the
present invention, while FIG. 11c shows a cross-sectional view
along a line 250 indicated in FIG. 11b through the device. The spin
device 100 comprises a more complex structure compared to the
previously described spin devices 100 in FIGS. 5 to 10. As a
result, the spin device 100 shown in FIG. 11 will be described with
references to the three FIGS. 11a, 11b and 11c at the same
time.
[0143] The spin device 100, once again, comprises an intermediate
semiconductor region 120 comprised in a substrate 110. The
substrate 110 may, for instance, comprise silicon (Si) or may be a
silicon substrate. The intermediate semiconductor region 120 abuts
the first terminal 130 (source terminal) and the second terminal
140 (drain terminal) which are aligned with the intermediate
semiconductor region 120 along a horizontally extending direction
in FIG. 11a. The contact for the control electrodes, which are not
shown in FIGS. 11a and 11b, extend along a line perpendicular to
the horizontal orientation indicated by arrows 800. As a
consequence, the cross-sectional view of FIG. 11c shows the control
electrode 190, but does not show the first and second terminals
130, 140 since the line 250 of the direction of the cross-sectional
view of FIG. 11c does not cross any of the two terminals 130,
140.
[0144] The cross-sectional view of FIG. 11c shows that the
intermediate semiconductor region 120 is formed as a mesa structure
with two sidewalls and a top surface. A magnesium oxide layer 400,
which is comprised in the spin selective scattering structure 170
(not labeled as such in FIG. 11c) so that along both sidewalls and
the top wall, interfaces 180 are formed by the deposited magnesium
oxide layer 400. The magnesium oxide layer 400 is, in turn, covered
by a ferromagnetic layer 410, which, in turn, is covered by a gate
oxide layer 420. As a consequence, both sidewalls and the top
surface of the mesa-like structured intermediate semiconductor
region 120 are covered by a stack comprising the sequence of a
magnesium oxide layer 400, a ferromagnetic layer 410 and a gate
oxide layer 420, such that the magnesium oxide layer 400 forms with
the sidewalls and the top surface of the intermediate semiconductor
region 120 and interfaces 180.
[0145] As outlined before, the ferromagnetic layer 410 may comprise
any of the previously mentioned ferromagnetic materials. In other
words, the ferromagnetic layer 410 may either be a conductive or
metallic ferromagnetic layer as well as a semi-conducting or
insulating ferromagnetic layer.
[0146] On top of the gate oxide layer 420, a control electrode 190
is deposited. FIG. 11c shows the case that only the top surface of
the mesa-structured intermediate semiconductor region 120 is
covered, at least partially, by the control electrode 190. However,
in different embodiments according to the present invention, which
is schematically illustrated in FIG. 11c by dashed lines, the
control electrode 190 may also cover a larger portion of the top
surface of the intermediate semiconductor region 120 and/or
partially or completely one or more sidewalls of the intermediate
semiconductor region 120. As a consequence, in the case of a spin
device 100 in the lateral design, up to three channel regions may
be formed inside the intermediate semiconductor region 120 close to
the interfaces 180 at the respective sidewalls and the top
surface.
[0147] Naturally, embodiments according to the present invention in
the form of the spin device 100 on the basis of the design shown in
FIG. 11 may also comprise biasing electrodes or floating gate
electrodes, which may be electrically insulated or in electrical
contact with the ferromagnetic layer 410, depending on certain
design and operational parameters. Therefore, spin devices 100
according to embodiments of the present invention may also comprise
more complex gate stacks.
[0148] FIG. 12a and FIG. 12b show a spin device 100 in the form of
a lateral device as already described in context with FIG. 11 with
multiple structures. The spin device 100, as shown in FIGS. 12a and
12b, comprises two separate intermediate semiconductor regions
120-1, 120-2, which abut each a first terminal 130-1, 130-2 and a
second terminal 140-1, 140-2.
[0149] Similar to the lateral device shown in FIG. 11, the control
electrodes 190, which are not shown in FIG. 12a, are connected
along a line perpendicular with respect to a line interconnecting
the first and second terminals 130, 140 of each of the two
parallel-shunted intermediate semiconductor regions 120. This line
is also indicated by the arrows 800 shown in FIG. 12a.
[0150] To simplify the description of the spin device 100 as shown
in FIG. 12a, a line 250 is shown in FIG. 12a, which indicates a
direction along which FIG. 12b shows a corresponding
cross-sectional view through the spin device 100.
[0151] The cross-sectional view of the spin device 100 illustrates
that the two intermediate semiconductor regions 120-1, 120-2 are
based on a substrate 110 and comprise mesa forms each. The
substrate 110 may, for instance, comprise silicon or may be a
silicon substrate. Each of the two mesa structures, therefore,
comprise two sidewalls and a top surface each.
[0152] The sidewalls and the top surface of the first intermediate
semiconductor region 120-1 are covered completely in the embodiment
shown in FIG. 12b by a first spin selective scattering structure
170-1. The first spin selective scattering structure 170-1
comprises a first magnesium oxide layer 400-1 and a first
ferromagnetic layer 410-1, which are deposited onto the mesa
structure of the intermediate semiconductor region 120-1 such that
along the sidewalls and along the top surface, three interfaces
180-1 are formed. Accordingly, the second intermediate
semiconductor region 120-2 is also covered by a second spin
selective scattering structure 170-2 comprising a second magnesium
oxide layer 400-2 and a second ferromagnetic layer 410-2, which are
deposited onto the mesa-structure of the second semiconductor
region 120-2 such that the second magnesium oxide layer 400-2 forms
three interfaces 180-2 along the sidewalls and the top surface.
[0153] The first and second spin selective scattering structures
170-1, 170-2 are then covered by a common or two distinct gate
oxide layers 420 that provide a lateral insulation of the
ferromagnetic layers 410-1, 410-2 with respect to control
electrodes 190-1, 190-2, which are deposited on top of the top
surface of the corresponding intermediate semiconductor regions
120-1, 120-2, respectively.
[0154] In the case of the FET-based implementation of the spin
device 100 shown in FIG. 12a and FIG. 12b, by applying
corresponding control voltages to the two control electrodes 190-1,
190-2, channel regions or channels may be formed during operation
in the vicinity of the interfaces 180-1, 180-2 at the top surfaces
of the mesas 120-1, 120-2.
[0155] However, the spin device 100 may not just be a parallel
implementation of two of the spin devices 100 as shown in FIG. 11a,
but the individual structures of the spin device 100 may differ
from one another. For instance, the spin selective scattering
structures 170-1, 170-2 may be adapted such that the two provide
charge carriers having different preferable spin orientations. In
the case of spin selective scattering structures 170 comprising a
magnesium oxide layer 400 and a ferromagnetic layer 410 as
schematically depicted in FIG. 12b, this may be achieved by
implementing the magnesium oxide layers with a different thickness
as outlined in the context of FIG. 4.
[0156] Based on the results previously described, the magnesium
oxide layer 400-1 of the first spin selective scattering structure
170-1 may, for instance, comprise a thickness of 0.2 nm (2 .ANG.),
while the magnesium oxide layer 400-2 of the second spin selective
scattering structure 170-2 may, for instance, comprise a thickness
of approximately 0.7 nm to 0.8 nm (7 .ANG. to 8 .ANG.). Naturally,
implementing the spin selective scattering structures 170-1, 170-2
in the described manner is, by far, not required. In principle,
implementing the spin selective scattering structures with the same
magnesium oxide layer thickness may also be realized in embodiments
according to the present invention.
[0157] As already outlined in the context of FIG. 11, FIG. 12 shows
a spin device with a fairly simply top gate stack. One or both
stacks provided on top of the intermediate semiconductor regions
120-1, 120-2 may furthermore comprise a floating gate electrode,
additional insulating layers, a biasing electrode or other
additional layers as previously described.
[0158] As the previous description of the spin device 100 shown in
FIG. 12 has shown, embodiments according to the present invention
may be implemented in such a way that a predetermined or "fixed"
spin-scattering property may result from the thickness of the
magnesium oxide layer 400 of the respective spin selective
scattering structure 170 only. Since it is possible to achieve
different spin polarization orientations based on varying the
thickness of the magnesium layer 400 alone without changing a
magnetization orientation of the ferromagnetic layer 410 deposited
on top of the magnesium oxide layer 400, embodiments according to
the present invention, as shown in FIG. 12, may be used as a spin
injector for injecting a current with a spin polarization, which is
different from an unpolarized current. Moreover, by using a spin
device 100 in the described fashion with two different spin
selective scattering structures 170, a switching between different
spin orientations may be possible without altering a common
magnetization or orientation of the ferromagnetic layers 410
involved.
[0159] Therefore, embodiments according to the present invention
may provide a selective spin injection, which may be controlled via
a control electrode (gate electrode) providing two different spin
directions, while having only one effective magnetization or
direction of the magnetic field active on the device. This may
become important in highly integrated circuits, where different
structures of the spin device 100 as shown in FIG. 12 or different
spin devices 100 are arranged closely to one another on the same
chip or die. In this case, by employing embodiments according to
the present invention, the ferromagnetic layers 410 of the spin
selective scattering structures 170 may be used having a common
orientation or a common magnetization direction. Compared to a
solution, which requires different magnetizations or different
directions of magnetizations for different spin injectors,
embodiments according to the present invention may significantly
reduce unwanted, parasitic interferences between the different
ferromagnetic layers.
[0160] Moreover, as the description of the previously mentioned
embodiments have already shown, by using an insulation layer as a
tunnel barrier between the control electrode, the ferromagnetic
material or a floating gate electrode, it may be possible to use
the spin device 100 in the mode of operation as a spin injector, as
a storage element, which provides, upon addressing (e.g., applying
a voltage to the first and second terminals 130, 140), a current
with a predetermined spin polarization. It may therefore be
possible to use the spin device 100 according to an embodiment of
the present invention also as a storage element for a spintronics
application.
[0161] As the description of the embodiments according to the
present invention shown in FIG. 12a has already shown, the
above-mentioned effects, challenges and demands may be fulfilled by
a spin device according to an embodiment of the present invention
that, for instance, comprise to individual metal-magnesium
oxide-semiconductor (drift). Each of these devices may employ a
first terminal (source region, source contact) and a second
terminal (drain region, drain contact), as well as some gate
contact on top of the magnesium oxide layer with a defined
thickness in between the two previously mentioned terminals.
[0162] FIG. 13a shows a perspective illustration of a spin device
100 comprising a first contact 150 (source metal contact), a
control electrode 190 (gate electrode) and a second contact 160
(drain metal contact). For the sake of simplicity only, the
terminals as well as the intermediate semiconductor regions are not
shown in FIG. 13a.
[0163] Between the first contact 150 and the second contact 160 a
spin selective scattering structure 170 is arranged. To be more
specific, in-between the first contact 150 and the second contact
160, the spin-scattering structure 170 is located, which comprises
a first magnesium oxide layer 400a directly in contact with the
semiconductor substrate 110 to form the interface 180 (not shown in
FIG. 13a), an optional second magnesium oxide layer 400b and a
ferromagnetic layer 410, which is also adapted to operate as the
respective control electrode 190. It should be noted that the
second magnesium oxide layer 400b is optional and may be, hence,
omitted if appropriate. It is not required to be present.
[0164] Concerning possible choices for materials and other related
properties and features, reference is made to the description of
the spin device 100 shown in FIG. 13b. A possible mode of operation
will also be described in the context of FIG. 13b.
[0165] FIG. 13b shows a perspective illustration of a proposed spin
device 100 comprising a common first contact 150 (source metal
contact), two control electrodes 190-1, 190-2 (gate electrodes 1
and 2) and two second contacts 160-1, 160-2 (drain metal contacts).
For the sake of simplicity only, the terminals as well as the
intermediate semiconductor regions are not shown in FIG. 13b.
[0166] Between the common first contact 150 and each of the two
second contacts 160-1, 160-2, spin selective scattering structures
170 are arranged. To be more specific, in-between the first contact
150 and the second contact 160-1, a first spin-scattering structure
170-1 is located, which comprises a magnesium oxide layer 400-1
directly in contact with the semiconductor substrate 110 to form
the interface 180 (not shown in FIG. 13b) and a ferromagnetic layer
410-1, which is also adapted to operate as the respective control
electrode 190-1.
[0167] A second spin selective scattering structure 170-2 is
located in between the first contact 150 and the second contact
160-2. In addition, the second spin selective scattering structure
170-2 comprises a ferromagnetic layer 410-2, which is adapted to
operate as the control electrode 190-2 of the respective structure.
However, different from the first spin selective scattering
structure 170-1, the second spin selective scattering structure
170-2 comprises two magnesium oxide layers 400-2a, 400-2b, wherein
the magnesium oxide layer 400-2a abuts the intermediate
semiconductor region 120 (not shown in FIG. 13b) to form the
corresponding interface 180 (not shown in FIG. 13b). The magnesium
oxide layer 400-2b is located in-between the magnesium oxide layer
400-2a and the ferromagnetic layer 410-2.
[0168] The spin device 100 shown in FIG. 13b is based on employing
a metallic ferromagnetic layer 410-1, 410-2, which is capable of
also operating as the gate electrode or control electrode 190-1,
190-2. Moreover, as indicated above, the common first contact 150
as well as the two second contacts 160-1, 160-2 are implemented as
metallic contacts, for instance, comprising aluminum (Al), copper
(Cu), titanium (Ti), tungsten (W), tantalum (Ta), gold (Au), silver
(Ag) or other metals and alloys. In different embodiments according
to the present invention, the first and second contacts 150, 160 as
well as other as metallic components described structures may also
be implemented based on a silicide (e.g., binary chemical compounds
comprising silicon (Si), e.g., TiSi.sub.2).
[0169] The first and the second spin selective scattering
structures 170-1, 170-2 along with the respective gate electrodes
190-1, 190-2 comprise different thicknesses of the magnesium oxide
layers in such a way that electrons or other charge carriers with
different spin orientations are scattered differently. While the
magnesium oxide layer 400-1 and the magnesium oxide layer 400-2a
may, for instance, be fabricated during the same fabrication
process steps having the same thickness, the magnesium oxide layer
400-2b, which is deposited onto the magnesium oxide layer 400-2a is
only deposited in the framework of forming the second spin
selective scattering structure 170-2.
[0170] To achieve the previously described different spin
scattering properties, which are also indicated by the
schematically-illustrated electrons 810-1, 810-2 in FIG. 13b, the
magnesium oxide layers 400-1, 400-2a may, for instance, comprise a
thickness of 3 atomic layers or monolayers. In contrast, the
magnesium oxide layer 400-2b comprised in the second spin selective
scattering structure 170-2 may comprise a thickness of 5 monolayers
or atomic layers.
[0171] Naturally, although in the embodiment shown in FIG. 13b the
ferromagnetic layers 410-1, 410-2 are adapted to directly work as
the control or gate electrodes 190-1, 190-2, also more complex gate
stacks may, once again, be employed comprising for instance
additional insulating layers.
[0172] Considering the operation of the spin device 100 as shown in
FIG. 13b, it is possible to address a path for the charge carriers
(e.g., the electrons 810) by addressing either of the two shown
metal-magnesium oxide-semiconductor structures by supplying
appropriate control voltages to the control electrodes 190-1,
190-2. It is therefore possible to turn off the spin device 100 by
supplying a corresponding voltage to both control electrodes 190 to
provide an unpolarized current by providing the control electrodes
190 with voltages such that at the second contacts 160-1, 160-2,
currents of equal magnitude and equal degree, but opposite spin
orientation, are provided. Moreover, control voltages may be
applied to the control electrodes 190-1, 190-2 that enable the spin
device 100 to provide a current with a controllable magnitude and a
controllable degree of spin polarization up to a value achievable
by a single spin selective scattering structure 170.
[0173] For instance, in the case of a spin device 100 based on
silicon with its very long spin-diffusion length of up to several
centimeters, it is possible not only to couple the first terminals
(not shown in FIG. 13b) of the two structures together, but also
the second terminals (not shown in FIG. 13b). By doing so, it may
be possible to achieve inside the semiconductor substrate 110, a
spin-polarized current with a controllable magnitude and a
controllable degree of spin polarization. Such a current may then
be further processed or influenced by further spintronic-related
devices.
[0174] Of course, apart from the mentioned magnesium oxide
insulating layers, materials other than magnesium oxide may be used
instead provided such materials have similar spin scattering
properties. Furthermore, in case the spin scattering properties of
the spin selective scattering structures 170 may be adjusted by
applying voltages to the control electrode or a bias electrode, an
embodiment similar to the one shown in FIG. 13b may also be
realized based on a single, individual metal-magnesium
oxide-semiconductor device. Naturally spin devices 100 according to
embodiments of the present invention may be based on a combination
of more than two such spin devices 100 as well.
[0175] FIG. 14 shows a further spin device 100 according to an
embodiment of the present invention, which is similar to the spin
device 100 shown in FIG. 13b. The proposed spin device structure as
shown in FIG. 14 differs, however, with respect to the control
electrodes (gate contacts), which are electrically insulated from
the ferromagnetic layers 410-2 by means of tunnel oxide barrier
420-1, 420-2.
[0176] In other words, on top of the previously described selective
spin-scattering structures 170-1, 170-2, an insulating layer 420-1,
420-2 are deposited, respectively, on top of which the control
electrodes 190-1, 190-2 are formed. As a consequence, it is
possible to use the ferromagnetic layers 410-1, 410-2 as floating
gate electrodes 430-1, 430-2 to achieve the previously described
programmability, since the spin device 100 shown in FIG. 14 may
offer the benefit in form of the option to use the ferromagnetic
layers 410-1, 410-2 as electron storage elements similar to
poly-silicon layers in flash storage cells. Hence, such a spin
device 100 comprising such a device structure may offer
programmable and/or a switchable spin injector that can be operated
in "off", "on", "spin-up" or "spin-down" modes, to name but a few.
In addition, the previously described mode of operation in which
the degree of spin polarization may be controllable, may be
implemented accordingly.
[0177] Moreover, the spin device 100 as shown in FIG. 14 may also
be used in the case of weakly conducting or non-conducting
ferromagnetic materials for the ferromagnetic layers 410-1, 410-2.
In this case, the ferromagnetic layers may not be usable as
floating gate electrodes 430 as denoted in FIG. 14. However, due to
the fact that the control electrodes 190 are still electrically
insulated from the ferromagnetic layers by the tunneling oxides or
gate oxide layers 420-1, 420-2, a negative influence caused by the
electrical charges provided to the control electrode 190 by
applying a corresponding control voltage on the ferromagnetic
layers 410-1, 410-2 may be prevented.
[0178] It should be noted that the figures presented so far
represent simplified illustrations of embodiments according to the
present invention, which are intended to facilitate a better
understanding of the great variety of possible spin devices 100
according to embodiments of the present invention. However, these
simplified illustrations may comprise inaccuracies such as
illustrations of layer thickness and lateral dimensions, which are
not to scale when compared to real-life implementations.
[0179] Moreover, it may be possible in some real-life
implementations according to embodiments of the present invention
that layers, such as insulating layers, may cover a wider or a
smaller range than depicted in the figures. It may also be possible
that some of the layers differ in terms of their extensions from
the figures. To name an example, it may be possible to limit the
extension of the control electrodes 190 to a smaller area compared
to the extensions of the spin selective scattering structures
170.
[0180] It should also be pointed out that the contacts as shown in
the figures are not required to be implemented. In other words, the
contacts may, for instance, be omitted when the respective
terminals (first terminals, second terminals) are directly
connected to further semi-conductive devices. The contacts (e.g.,
first contacts, second contacts, control contacts) may be
implemented on the basis of semi-conductive materials, metals or
other conductive elements, compounds and materials.
[0181] While in the previous sections of the specification the
focus was mainly on generating spin-polarized current, the spin
devices 100 according to embodiments of the present invention may
also be used as spin detectors or detectors for determining a spin
polarization. Since all the spin devices 100 described so far are
based on the presence of spin selective scattering structures 170,
which provide differing scattering cross sections for charge
carriers with different spin polarizations, single spin devices 100
as well as more complex circuitries comprising one or more spin
devices 100 may be used as spin detectors.
[0182] Due to the different cross sections for scattering charge
carriers depending on their spin orientation, the spin devices 100
may comprise changing resistance values depending on the spin
polarization and/or the degree of spin polarization of a current
provided to the first terminal of these devices.
[0183] Without providing a feedback signal or a control signal to
the control electrodes 190 of the spin devices 100, a single spin
device 100 may be used to detect the spin polarization of a current
by measuring its resistance value.
[0184] A further spin detector 900 according to an embodiment of
the present invention employing two spin devices 100-1, 100-2 is
schematically shown in FIG. 15a. The spin detector 900 comprises an
input 910, to which the current, the spin polarization of which is
to be detected, is provided. The input 910 is coupled in parallel
to the two spin devices 100- 1, 100-2, which are coupled with their
first terminals 130-1, 130-2 to the input 910. A high input
impedance voltmeter 920 or a high input impedance electrometer 920
is coupled to the second terminals 140-1, 140-2 of the two spin
devices 100-1, 100-2.
[0185] The two spin devices 100-1, 100-2 are adapted to provide
different spin polarizations. For instance, the spin device 100- 1
may be adapted to provide charge carriers, when operated as a spin
injector, with the opposite spin orientation compared to the spin
device 100-2. As a consequence, the two spin devices 100-1, 100-2
will scatter charge carriers provided to their first terminals
130-1, 130-2 differently.
[0186] Due to these different scattering cross sections for the
different spin polarizations of the charge carriers provided to the
two different spin devices 100-1, 100-2, at the second terminals
140-1, 140-2 of the spin devices 100-1, 100-2, different
electro-chemical potentials may be present. The difference between
the two different electro-chemical potentials may then be detected
by the high input impedance voltmeter 920. Therefore, a voltage
difference between the two second terminals 140-1, 140-2 of the two
spin devices 100-1, 100-2 represents a measure for the spin
polarization of the current originally provided to the two first
terminals 130-1, 130-2.
[0187] FIG. 15b shows a further spin detector 900' according to an
embodiment of the present invention with an input 91 0. However, in
contrast to the spin detector 900 shown in FIG. 15a, the spin
detector 900' comprises a single spin device 100, which is coupled,
with its first terminal 130, to the input 910. The spin detector
900' further comprises an Ohmic resistor 930, which is coupled in
series with a second terminal 140 of the spin device 100. A
differential amplifier 940 is coupled with its inputs to the two
terminals of the resistor 930 and, with an output, to the control
contact 200 of the spin device 100. Moreover, an Ohm-meter 950 is
coupled in parallel to the spin device 100.
[0188] Providing a current with a spin polarization to be
determined to the input 910 will result in a resistance value of
the spin device, which depends on the spin polarization of the
current. However, since the Ohm-meter 950 typically comprises high
input impedance values, the current will reach the resistor 930 and
create a voltage drop across the resistor 930, which is then
detected by the differential amplifier 940. Although the spin
polarization may be changed due to the interaction with the spin
device 100, the magnitude of the current through the spin device
100 will remain unchanged.
[0189] The voltage drop across the resistor 930, which is detected
and optionally further processed by the differential amplifier 940,
is a measure for the magnitude of the current provided to the input
910. Due to the output of the differential amplifier 940 being
coupled to the control contact 200 of the spin device 100, a
feedback loop is generated so that the magnitude of the current
flowing into the spin device 100 will be altered to be independent
of the degree of spin polarization of the current provided to the
input 910. As a consequence, a resistance value measured by the
Ohm-meter 950 across the spin device 100 is a measure for the
degree of spin polarization of the current provided to the input
910.
[0190] While the foregoing has been particularly shown and
described with reference to particular embodiments thereof, it will
be understood by those skilled in the art that various other
changes in the form and details may be made without departing from
the spirit and scope thereof. It is to be understood that various
changes may be made in adapting to different embodiments without
departing from the broader concept disclosed herein and
comprehended by the claims that follow.
* * * * *